Drummond Reed wrote: >>> Dick Hardt wrote: >>> >>> :-) ... that label would be more accurate. There is lots of work to be >>> done to make OpenID simpler for users. I think that what will be easy >>> for users is something provided by the browser that lets the user >>> click to initiate a login or registration. No typing is better then >>> any typing! Back when we started working on the protocols we could not >>> expect this kind of functionality to be in the browsers. Now that >>> awareness is higher, having it built into the browser is feasible. I >>> of course am biased given the work we have done with Sxipper >>> http://sxipper.com :) >>> >> For the majority of users, this is probably the most likely path of >> introduction to OpenID. Note that it's not just about allowing the user >> to do something with one click, but also about being proactive and >> informing the user that they can login to a site with an identity they >> already have. This can be as simple as telling the browser "identity >> agent" (e.g. sxipper) which email addresses the user has and letting the >> identity agent figure out which OpenID's the user has that they don't >> even know about. >> >> George Fletcher wrote: >> >> I think relying party sites that support OpenID could do more to make it >> clear on their home pages that they support OpenID (as often it's hidden >> behind another click). This could be as simple as some <link> tags that >> advertise support for OpenID. Maybe a <link> to the XRDS doc describing >> the services of the site. Then the identity agent can discover the >> relying party OpenID return_to endpoint and log the user in directly. >> Can be used to solve a phishing problem and makes the experience easy >> for the user. >> >> Some related thoughts .... >> http://practicalid.blogspot.com/2007/06/clients-to-rescue.html >> >> http://practicalid.blogspot.com/2007/06/passive-identity-meta-system- >> markup.html >> > > George, I read your two posts with great interest...and then noticed that > they were last summer! > > You are a man ahead of your time. > > Where has discussion of your "IDMML" gone since your posts? > > =Drummond Unfortunately, not as far as I'd like :( I've not been able to get back to the ideas and take them farther. With the other things that have happened in the last 6 months there are needed revisions. Maybe this could be a discussion at IIW (if there is enough interest)?
At the time there was less consensus around XRDS as a service "description/meta-data" markup. With that changing, the time is better to move this forward. I suspect there are significant synergies with what Peter hinted at in the work with XRDS, IDP Discovery, and SAML. It would be great if identity agents could be the glue that binds the different identity systems together for the user (until we on the technology side get closer to real convergence:). Thanks, George _______________________________________________ specs mailing list specs@openid.net http://openid.net/mailman/listinfo/specs
